Linking Machine Learning with Multiscale Numerics: Data-Driven Discovery of Homogenized Equations.
Hassan ArbabiJudith E. BunderGiovanni SamaeyAnthony J. RobertsIoannis G. KevrekidisPublished in: CoRR (2020)
Keyphrases
- data driven
- multiscale
- machine learning
- knowledge discovery
- data mining
- wavelet transform
- learning systems
- image representation
- scale space
- image segmentation
- decision trees
- numerical solution
- mathematical model
- natural images
- knowledge acquisition
- image processing
- partial differential equations
- feature selection
- differential equations
- computer science
- data analysis
- data sets
- discovery process
- text mining
- supervised learning
- learning algorithm
- active learning
- linear systems
- data preparation
- inductive learning
- inductive logic programming
- computer vision
- machine learning methods
- multiresolution
- edge detection
- information extraction
- reinforcement learning
- pattern recognition
- shape representation
- machine learning approaches
- model selection
- linear equations
- nonlinear equations
- multiscale analysis